That HIV/AIDS is a serious problem plaguing Africa is essentially a truism; commonly accepted and virtually undisputed in establishment media or press. And the unofficial reason for the severity of the problem, namely the over-sexualized behavior of blacks, while not a truism, does obviously have some rather widespread appeal even in intellectual culture. But could a commonly accepted truth in fact turn out to be little more than gross misunderstanding of appropriate scientific methodology? Could the prevalence of HIV/AIDS in South Africa have more to do with flawed statistical analysis than with the facts?
Aids in South Africa: The Medical Science of Racial Apartheid
In the Contemporary World Regional Geography text, the authors state that:
The causes of high levels of HIV/AIDS in Africa include poverty, the breakdown of traditional family support systems, the apartheid policy in South Africa that brought miners into male-only camps serviced by prostitutes, continuing promiscuity at a time when traditional polygamy gives way to the taking of sexual partners outside monogamous marriages, and mistaken government policies. HIV/AIDS spreads quickly in cultures that value male sexual practices. (Bradshaw et al. 2007: 398)
What is meant by all this is quite clear in less polite terms, that HIV/AIDS is a problem in South Africa because of the promiscuous and irresponsible sexual behavior of blacks. That HIV/AIDS is a serious problem plaguing Africa is essentially a truism; commonly accepted and virtually undisputed in establishment media or press. And the unofficial reason for the severity of the problem, namely the over-sexualized behavior of blacks, while not a truism, does obviously have some rather widespread appeal even in intellectual culture.
As with any idea that has become commonly accepted, the acceptance of the underlying reality makes the general public and establishment academic/scientific community willing to accept the statistics that emerge, without scrutiny, as a description, rather than a justification of material facts. But could a commonly accepted truth in fact turn out to be little more than gross misunderstanding of appropriate scientific methodology? Could the prevalence of HIV/AIDS in South Africa have more to do with flawed statistical analysis than with the facts?
In looking at the statistical data provided by on the website of a prominent AIDS charity, the section on South Africa emphasizes two studies, the former a "report of the Department of Health 'National HIV and Syphilis Sero-prevalence Survey in South Africa 2006', published in 2007" is featured (http://www.avert.org/safricastats.htm). This report is the "17th in a series of annual studies which look at data from antenatal clinics and use it to estimate HIV prevalence amongst pregnant women." This study has a sample with a number of cases equal to 33,033, the cases being women receiving care at the 1,415 antenatal clinics. These data are divided into the nine provinces and show that 29.1% of clinic attendees were estimated to be infected with the disease (down from 30.2% in 2005) which affords the Avert the confidence to declare, quite precipitously, that "29.1% of pregnant women were living with HIV in 2006" (Ibid.).
The latter study, a "report of the 'South African National HIV Prevalence, HIV Incidence, Behaviour and Communication Survey, 2005.'" In this study a non-random "sample of people were chosen to represent the general population, 55% of whom agreed to give a blood sample to be anonymously tested for HIV." These national HIV prevalence survey data were collected by...
...fieldworkers (who) visited 12,581 households across South Africa, of which 10,584 (84%) took part in the survey. Of the 24,236 people within these households who were eligible to take part, 23,275 (96%) agreed to be interviewed and 15,851 (65%) agreed to take an HIV test. This means that only 55% of eligible people were tested. (Ibid.)
This report showed a 10.8% prevalence of HIV in all South Africans over the age of 2 in the year the study was conducted, 2005. Notably much lower than the roughly 30% of pregnant women who were estimated to be infected. A hard-and-fast conclusion could be drawn that it is the poor pregnant (sexually active) women who must bear the brunt of the HIV crisis. However, that conclusion may be a bit premature.
The Department of Health HIV prevalence survey tested only one sample group, women visiting antenatal clinics (meaning pregnant women) which is a very narrow demographic range. The survey showed that 29.1% of women visiting the antenatal clinics were infected with HIV, which is consistent with the data from the United Nations and Population Reference Bureau presented on page 401 of the Bradshaw text which illustrates that 20-29% of the adult population of South Africa according to 2003 data. Though this 20-29% is quite a differential from the 10.8% estimated in the National HIV prevalence survey of the general population. Perhaps the UN are relying primarily on samples from antenatal clinics, which is a huge mistake.
This is a huge mistake because HIV testing in antenatal clinics has been demonstrated to result in a large number of false positives, particularly when rapid serum screening techniques are used (http://archfami.ama-assn.org/cgi/content/full/9/9/924). This would be perfect if one were intentionally trying to inflate the ostensible prevalence of HIV in South Africa. But of course why would one want to do such a thing? A 2005 study published in The Lancet shows that pregnancy increases a woman's chance of getting HIV by 100%, even when controlling for sexual behavior (http://www.aidsmap.com/en/news/E4405364-04D0-4F58-83F1-C0FD43653AA3.asp). This is an interesting proposition which can be stated by saying that the very nature of being pregnant increases a woman's risk for HIV infection. Yet, this author has not seen any empirical data systematically showing that being pregnant is the causality of increased risk for HIV independent of variances in sexual behavior.
According to a news report "Rapid HIV tests: false positive risk without confirmatory test highlighted in Uganda," highlighted a study from the British Medical Journal which demonstrated that 86% of "weakly positive" HIV antibody screens from rapid diagnostic tests turned out to be false positives when more reliable confirmatory tests were done (http://www.aidsmap.com/en/news/EED7F124-74A6-41C6-94BA-BCB27549E00A.asp). The study went on to say that, "With increased access to antiretroviral therapy (ART) being the target for many resource-poor countries lacking good laboratory facilities, the use of rapid diagnostic tests will be scaled up to an even greater level." This means that as the HIV pharmaceuticals industry becomes more entrenched in impoverished nations, the incidence of HIV false positives generated by accepted screening parameters will increase dramatically, which is not counter-intuitive.
So not only does prenatal screening lead to a high volume of false positive and indeterminate results, but so do rapid serum screening techniques. However, this is not a deviation from the standard of care utilized in HIV preventive medicine in Africa. Twenty years ago World Health Organization "authorities decided HIV tests were so expensive doctors didn't need them to diagnose AIDS in Africa. Patients with 10 percent weight loss, persistent fever and diarrhea and a cough became AIDS cases by definition" (Bethell 2005). Though rapid weight loss, fever, diarrhea, a cough and a weak immune response are all symptoms of malnutrition, which is a common problem in the poorest continent on earth (Antartica excluded). They are also symptoms of other diseases of the poor, such as tuberculosis.
What is the economic impact of HIV/AIDS? Well, for South Africa it is a presumed reduction in the workforce. But for for corporations it is the major expansion of one industry: pharmaceuticals. "Pharmaceuticals are a growing area for generic drugs that the government can buy cheaply, especially the anti-retroviral protection for HIV/AIDS sufferers" (Bradshaw 2007: 433).
Bethell, Tom.Nov 15, 2005 "Pandemic jitters." The Washington Times: (DC)
Bradshaw, Michael et al. 2007. Contemporary World Regional Geography: Global Connections, Local Voices Second Edition. New York: McGraw-Hill.